Jim Moy

Over the years, there has been discussion on the Internet about origin of the name of the software protocol known as TWAIN, a standard allowing diverse applications to connect with sources of image data like scanners. Nobody seems to know exactly why it was given its name. Even the standards body that created it (and ought to know) references the Free Online Dictionary entry with no further comment.

The definition there is largely accurate, but incomplete. Because no official acronym was chosen, confusion occurs as people make up their own meanings and then assert or argue it based on hearsay. Amusingly, I saw the issue discussed in the alt.battlestar-gallactica Usenet groups as recently as Oct 2005.

As I was present in the meeting where the name TWAIN was selected for the standard, I thought I might add some background to the issue for the curious. Not that I think it’s a particularly interesting discussion. But the fact that the name is in capital letters combined with nobody knowing for sure why there is no official acronym seems to produce a sort of cognitive dissonance where the void requires filling by whatever means available.

HP Greeley R&D

So. Back in 1992 the meeting took place in a Dilbert-style cubicle at Hewlett-Packard’s Greeley, Colorado office, the source of the original ScanJet line of scanners. HP was involved in the standard’s working group committee, and the cubicle was occupied by Kevin Biers, an R&D engineer who was serving as HP’s representative to the committee, and who had been serving as its chairman. Also present was Jim Graham and myself, engineers in the same R&D lab, and working on various projects related to scanners, back in the day when they were designed and manufactured in one building with no overseas assistance.

It was more of a gripe session than an official meeting. Jim and Kevin were lamenting news from corporate legal that the most recent in a string of rejected standards names, “Direct Connect,” was also unavailable due to trademark conflicts. This was back in the day when you couldn’t just Google a name to see whether it was being used commercially. I was present only because my cubicle was directly across the divider from Kevin’s and I was investigating an expletive followed by a slam of a telephone handset into its cradle.

There was a deadline for a name because the initial PR for the standard was being prepared, to be introduced in Byte magazine, at the time a popular rag for the personal computer crowd. Teresa Simske, a colleague at HP, was doing the preparation and had been asking Kevin what the standard should be called. With a hint of exasperation, Kevin commented that he was going to have to call her back and say that she would have call it “the spec without a name,” referring to the “specification” — the document which described the exact functioning of the standard, down to the API level.

Refinement

A split-second later, either Kevin or Jim (I can’t remember precisely which, or in what order because of the commotion it caused) lit up and said “The spec without an *interesting* name!” and “Hey, Twain, then it’s actually a word!” Hilarity ensued, with references to Mark Twain and the Kipling poem, and with the winking acknowledgement that it would be fun to take an absurd chain of events like this and get to put it into what would otherwise be a dry, technical document. The 1.0 version of the standard was almost completed, and there was some punchiness about this because of the irony of doing all that work, and having the most pressing issue being the bestowing of a name.

I didn’t participate in the phone calls and whatever other activity it took to make the name official. Presumably the other members of the working group were also weary of the name search and so did not take much convincing. Like many people who successfully chair standards committees, Kevin had good people skills and so I guess that had no difficulty convincing them. Another detail is that Kevin recounted a phone conversation with Teresa about the unwieldiness of the name, and they came to agreement that the word “spec” was not important, and so “Technology” would replace “The spec” to produce Technology Without An Interesting Name for the PR release.

Toolkit

There have also been references to “Toolkit” instead of “Technology” Without an Interesting Name as well. One example is this post from Bob Gann, who was also an R&D engineer on the hardware side of the original ScanJet projects:

The word “toolkit” refers to the various software components that accompanied the specification, with reference source implementations, C programming language headers, example programs, etc. I remember the term being used within the lab in the months after the naming incident — note the date of the post — and so one might interpret it as an equally valid expansion of the acronym.

And that’s the story, for what it’s worth. So the acronym does “haunt” the standard as FOLDOC says, but only because the working group doesn’t state there existed an actual acronym, or is hesitant to adopt it (grin).

And Who are You?

The reader might ask what qualification I have to state this information authoritatively, to which I can only answer, “I was there,” which has its limits. I went on to author the original Macintosh TWAIN source for the HP ScanJet series that served as an early implementation reference, as well as performed application side implementation for the Macintosh version of Adobe Photoshop. Mike Niquette, another HP engineer performed similarly on the Windows side. Later, I authored the ScannerBe standard for the now defunct BeOS. Jim Graham, Kevin Biers, and Teresa Simske have gone on to other things, both in and out of HP. Having left HP in 1993, I can’t comment on whether the FOLDOC description about an acronym contest actually occurred, but it hadn’t to my knowledge before then. Some other engineers I knew at HP are still there, I wonder what they might say about it?

Cross-posted to the Wikipedia.

Time to learn something new. I’ve been through the tutorials on both Ruby on Rails and Seaside. I’m not sure to which I’m going to devote my limited spare time outside of work. Not enough bandwidth for both, at least not at the same time. Rails is all the rage, and Ruby is great, but it’s still the familiar MVC and DB kind of thing. Seaside is a complete brain-inside-out experience, so I’m leaning that way…

ROTFL

http://www.artima.com/weblogs/viewpost.jsp?thread=74230

“…coding style is an essentially solved problem, and we ought to stop worrying about it. And to stop worrying about it will require worrying about it a lot first, because the only way to get from where we are to a place where we stop worrying about style is to enforce it as part of the language.”

“Perl is a vast swamp of lexical and syntactic swill and nobody knows how to format even their own code well, but it’s the only major language I can think of … that doesn’t have at least one style that’s good enough.”